| Project/Area Number |
25380481
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Research Field |
Management
|
|---|
| Research Institution | Tokyo City University |
|---|
Principal Investigator |
Umehara Eiichi 東京都市大学, メディア学部, 教授 (00645426)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SUWA Hirohiko 奈良先端科学技術大学院大学, 情報科学研究科, 助教 (70447580)
OGAWA Yuki 立命館大学, 情報理工学部, 助教 (40625985)
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000)
Fiscal Year 2015: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2014: ¥1,170,000 (Direct Cost: ¥900,000、Indirect Cost: ¥270,000)
Fiscal Year 2013: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | ソーシャルメディア / 株式市場 / 自然言語処理 / 機械学習 / 経営情報 / ファクターモデル / 株式予測モデル / 株式掲示板 |
|---|
| Outline of Final Research Achievements |
We analyzed the relationship of Yahoo! stock BBS and the stock market. The First, I found a new factor for explaining the stock returns. We calculated the term frequency (TF / IDF) using a natural language processing, and extracted factors using a principal component analysis. As a result, we could extract 3 factors (the investment strategy, the stock holdings, and the financial economy). The second, we analyzed the relationship between Nikkei Stock Average Volatility Index (VIX) and Yahoo! Stock BBS. We applied a topics analysis (Latent Dirichlet Allocations) to messages in the stock BBS. We found a correlation between the posted rate for each topic and VIX. The Third, we analyzed a relation between Twitter and Viewing rate in midnight TV anime programs using our method developed in this study. We found that there is a significant correlation between the number of messages posted on Air and the number of negative words in the tweets on Air, and next week viewing rates.
|
